Nvidia Hit with False Advertising Suit over GTX 970 Performance

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Kenmitch

Diamond Member
Oct 10, 1999
8,505
2,249
136
I wish NVidia would get fried. I hate this no ethics, anything for money trash. I can't believe people are ok with them lying. NVidia knew exactly what they did.

NVidia has a cult like following. The members will defend them till the end. Not sure why they feel this way at all.

AMD can hit a grand slam on launch and the cult will come out full swing moving the bars to meet their agenda.

Guess the cult has the Pinky and the Brain syndrome....Go figure!

Warning issued for trolling.
-- stahlhart
 
Last edited by a moderator:

Headfoot

Diamond Member
Feb 28, 2008
4,444
641
126
ITT posters flail about claiming false advertising, apparently unaware there are 3 other causes of action in the complaint, and further unaware that it's not a Lanham act "false advertising" claim -- it is a california state law misleading advertising claim. Funny that nobody has bothered to notice the (highly important) differences there after 4 pages of back and forth
 

Pandamonia

Senior member
Jun 13, 2013
433
49
91
ITT posters flail about claiming false advertising, apparently unaware there are 3 other causes of action in the complaint, and further unaware that it's not a Lanham act "false advertising" claim -- it is a california state law misleading advertising claim. Funny that nobody has bothered to notice the (highly important) differences there after 4 pages of back and forth

Also it is not false advtertising.

Its a type/mistake in one number.

they didnt come out saying that GPU would get you laid more or make you better looking.

I dont see what damages people can claim since they didnt lose anything. People cannot even say what performance difference there would be with less ROP's and under what conditions.
 

oobydoobydoo

Senior member
Nov 14, 2014
261
0
0
Also it is not false advtertising.

Its a type/mistake in one number.

they didnt come out saying that GPU would get you laid more or make you better looking.

I dont see what damages people can claim since they didnt lose anything. People cannot even say what performance difference there would be with less ROP's and under what conditions.


How is not false advertising? If I want to play titanfall with full details and textures, and the requirements say "4GB VRAM"... and I buy a GTX970... what happens when I try to play titanfall with full textures?


Either the nvidia driver recognizes the game, and tricks it into lowering the details to use less than 3.5GB for it to work properly.... OR the game becomes a slideshow because the 970 actually cant use the full 4GB of Vram in to render the freakin' game!!


The card can't run the game the way a 980 can, but it claims it can on the box. That's false advertising. It cites a 256bit memory bus, but it actually can't use it. It cites 64 ROP... but only has 56 available to the game engine. That's BS.
 
Last edited:

oobydoobydoo

Senior member
Nov 14, 2014
261
0
0
ITT posters flail about claiming false advertising, apparently unaware there are 3 other causes of action in the complaint, and further unaware that it's not a Lanham act "false advertising" claim -- it is a california state law misleading advertising claim. Funny that nobody has bothered to notice the (highly important) differences there after 4 pages of back and forth

ITT You show how biased you really are.
 

Spanners

Senior member
Mar 16, 2014
325
1
0
Also it is not false advtertising.

Its a type/mistake in one number.

they didnt come out saying that GPU would get you laid more or make you better looking.

I dont see what damages people can claim since they didnt lose anything. People cannot even say what performance difference there would be with less ROP's and under what conditions.

I don't think (assuming Nvidia's explanation is accurate) that the cause of the incorrect specifications have any relevance. If I was selling a 64g chocolate bar that turned out to only be 56g then the fact that it was a typo or the factory failed to pass on the correct weight information to the labeling department doesn't matter it's still false advertising.

Regardless it was more than one number they misrepresented the L2 cache as well as the ROP count. You could make an argument in regards to the bandwidth numbers, bit width of the memory interface and the 80% slower segment of the memory also.

"As a result of Defendants’ “unfair” business practice, Plaintiff and members of the Class spent money on the GTX 970 devices that they would not otherwise have spent at the amount charged by Defendants and did not receive the capabilities promised by Defendants."

People can definitely show cases were the performance is affected under certain conditions.
 

Pandamonia

Senior member
Jun 13, 2013
433
49
91
I don't think (assuming Nvidia's explanation is accurate) that the cause of the incorrect specifications have any relevance. If I was selling a 64g chocolate bar that turned out to only be 56g then the fact that it was a typo or the factory failed to pass on the correct weight information to the labeling department doesn't matter it's still false advertising.

Regardless it was more than one number they misrepresented the L2 cache as well as the ROP count. You could make an argument in regards to the bandwidth numbers, bit width of the memory interface and the 80% slower segment of the memory also.

"As a result of Defendants’ “unfair” business practice, Plaintiff and members of the Class spent money on the GTX 970 devices that they would not otherwise have spent at the amount charged by Defendants and did not receive the capabilities promised by Defendants."

People can definitely show cases were the performance is affected under certain conditions.

What performance is promised?

I have yet to see nvidia or amd ever promise anything. half the time you cannot find out if the GPU is fully activated or what is disabled. it mostly comes from review sites rather than them directly

Does the box say how many ROPS the card has? Does Nvidias advertising? or are we using slides from some presentation or some websites spec sheets?
Where is Nvidias direct marketing saying this is such?

Tell me what performance has been lost? how can this be quantified?

UK Retailers show CUDA Cores and not ROPS on the adverts....
 
Last edited:

AnandThenMan

Diamond Member
Nov 11, 2004
3,949
504
126
^^ all of that and more has been covered already. If you want to know all the details read the filing it's all in there.
 

kevinsbane

Senior member
Jun 16, 2010
694
0
71
What performance is promised?
Not an issue of promise of performance. Misleading specs - using the same specs for a GTX 970 and a GTX 980, when there are material differences in memory speed.

Tell me what performance has been lost? how can this be quantified?
Straight from the horse's mouth:
PcPer and Nvidia said:
NVIDIA’s performance labs continue to work away at finding examples of this occurring and the consensus seems to be something in the 4-6% range. A GTX 970 without this memory pool division would run 4-6% faster than the GTX 970s selling today in high memory utilization scenarios.

That's putting aside the fact that I would not have bought it if I had known. So my loss is the cost of the card.
 

Creig

Diamond Member
Oct 9, 1999
5,171
13
81
Also it is not false advtertising.

Its a type/mistake in one number.

they didnt come out saying that GPU would get you laid more or make you better looking.

I dont see what damages people can claim since they didnt lose anything. People cannot even say what performance difference there would be with less ROP's and under what conditions.

Mistake or not, it is up to Nvidia to ensure that the specs they release match the capabilities of the card.

Nvidia stated 64 ROPs. Actual ROPs = 56
Nvidia stated 2MB L2 cache. Actual L2 cache = 1.75MB

People who purchased a 970 did not receive a card that matched the specs that were released by Nvidia. Thus the class action suit. Nvidia doesn't have a leg to stand on here.
 
Last edited:

Spanners

Senior member
Mar 16, 2014
325
1
0
What performance is promised?

I have yet to see nvidia or amd ever promise anything. half the time you cannot find out if the GPU is fully activated or what is disabled. it mostly comes from review sites rather than them directly

Does the box say how many ROPS the card has? Does Nvidias advertising? or are we using slides from some presentation or some websites spec sheets?
Where is Nvidias direct marketing saying this is such?

Tell me what performance has been lost? how can this be quantified?

UK Retailers show CUDA Cores and not ROPS on the adverts....

Nobody has claimed Nvidia failed to deliver on promised performance or even that any promises were made. I made no mention of it either.

Nvidia's reviewer's guide was the source of the incorrect information, in fact there are still links on Nvidia's website that lead to reviews with these specification errors. Here, here and here. All link directly from the 970 page on Geforce.com using information provided directly by Nvidia.

"It's not really about what Nvidia has decided to cut from the second-tier 900-series card, it's about what they've left in.

The GPU is still running with the same 2MB of L2 cache onboard, is still rocking the same sixty-four render output units (ROPs) and still retains the same algorithmic advances that makes the 256-bit memory bus viable for a card that's aiming at 4K gaming performance. And it's the 4GB of GDDR5 video memory Nvidia have retained for their second-tier Maxwell card that makes it still so competitive despite being priced far lower than it's big brother.
"

That's the kind of statement that would encourage me to purchase a card and it's factually incorrect.

The performance loss can be quantified by increased frame-time latency when a games VRAM requirements pass 3.5GB.
 
Last edited:

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
People can definitely show cases were the performance is affected under certain conditions.

At this point in time I have yet to see a proper statistically correct quantitative analysis from any review site.

That's putting aside the fact that I would not have bought it if I had known. So my loss is the cost of the card.

That is what you say now. Or were you one of the first to buy a 970 (when it was priced really well) and would still have bought it if you had seen the correct specs.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
At this point in time I have yet to see a proper statistically correct quantitative analysis from any review site.



That is what you say now. Or were you one of the first to buy a 970 (when it was priced really well) and would still have bought it if you had seen the correct specs.

Interesting that all reviews have virtually stopped on the 970. I say it just shows that the tech press really isn't concerned with giving us the answers. Just selling product for the industry and collecting their paychecks.

He said he would not have bought it. Why ask him again? This is a "read my lips" moment.
 

garagisti

Senior member
Aug 7, 2007
592
7
81
I wish NVidia would get fried. I hate this no ethics, anything for money trash. I can't believe people are ok with them lying. NVidia knew exactly what they did.
Fool me once, shame on you. Fool me twice, shame on me.

Nvidia is also reported to have advertised the specs on their website. Clearly it isn't just a mistake in the reviewers guide as i thought 1 review site suggested. It is a classing example of bending over and asking for more punishment from the review sites, who should technically lean towards their readership more than anything, as no readership equals no hits, ergo no advertising money.

I still can't believe all the ferocious defending on this cheap trick that consumers were had with. The same with OC on mobile parts, and yet more defending.

It will be fun to read what people may say after the verdict/ settlement. People mocked about bumpgate too, and that was painfully real.
 

garagisti

Senior member
Aug 7, 2007
592
7
81
At this point in time I have yet to see a proper statistically correct quantitative analysis from any review site.



That is what you say now. Or were you one of the first to buy a 970 (when it was priced really well) and would still have bought it if you had seen the correct specs.
People were buying/ recommending buys (i recommended a few and was considering buying before this became public) given it was reflecting specification of 980 and not that far off 980 and better than 290s in price/ performance at launch. However, imho, if you think that actual knowledge of the weird memory configuration will not have killed sales of 970, you probably don't belong on tech forums. Heck, Nvidia users like GoldenTiger(at XS and HardForum), who don't use anything but Nvidia are reporting a problem running SLI at 1440p with them, and that guy is a loyalist who has no reason to lie. So yes, people will have been more actively suggesting to stay away from 970 unless you needed Nvidia specific features and budget was price of 970 and so on. On the other hand, it will have led to higher sales of 980 in the beginning, but overall Nvidia definitely will have seen lower sales for 980 and 970 combined, and sale of 970 outright.
 

kevinsbane

Senior member
Jun 16, 2010
694
0
71
At this point in time I have yet to see a proper statistically correct quantitative analysis from any review site.
There's a quote from Nvidia yourself. 4-6% penalty under high memory use. And frankly, their analysis is worth more, since they most likely actually have access to a real 64 ROP/2mb L2 Cache GTX 970. All the rest of us can do is speculate wildly about the theoretical performance of a fully enabled memory GTX 970.


That is what you say now. Or were you one of the first to buy a 970 (when it was priced really well) and would still have bought it if you had seen the correct specs.
Or - I really would not have bought it based on the real specs. Frankly, I'm going to trust my own judgement about my own actions, not your insinuations about what I would or would not have done.

In fact, no need to guess. I pestered my retailer, manufacturer and NVIDIA themselves for days. I got my refund, didn't settle for just the MSRP - i wanted my taxes and shipping back too.

And even supposing that I myself would have bought the card anyways, (a most dubious and unfounded assumption), the point of my post is still the same - the fact that there exists a class of persons (of which I count myself amongst) for whom the recent revelations about the GTX 970 would have dissuaded them from buying it - thus, the loss to those people is the cost of the card. If someone asks "what possible loss could there be?", well, there's at least one answer.
 
Last edited:

rgallant

Golden Member
Apr 14, 2007
1,361
11
81
I still wonder about the gm200 cut downs and if they were\will have the same split memory pools ?
to bin more chips with 1 or 2 L2 defects

should be fun at the first card reviews ,

per law suit I wonder if they will pull the nv drivers apart to see how it treats the 980 vs 970 memorys to show they are not the same.
funny if they find a 3.5 gb soft cap for the 970.
 
Last edited:
Aug 11, 2008
10,451
642
126
Not an issue of promise of performance. Misleading specs - using the same specs for a GTX 970 and a GTX 980, when there are material differences in memory speed.

Straight from the horse's mouth:

That's putting aside the fact that I would not have bought it if I had known. So my loss is the cost of the card.

Well, if the whole cost of the card is lost, just send it to me, since it is now worth nothing.

Seriously, you can argue that some performance is lost, and some specs are overstated, but the card is hardly useless, just gives (slightly?) less performance under some circumstances, but even that is very hard to quantify. The most that a buyer can say is "lost" is the difference in performance between the 970 and what a competing card of the same price would have given if you had not bought the card (the 970).

Make no mistake. I think nVidia did a bad thing here. Either they made a legitimate mistake and tried to cover it up, or they were deliberately deceptive from the beginning. In any case, they probably deserve come kind of penalty. I just think some of the outrage and hysteria over this has gotten out of control. It is not like a defective car that kills people or a drug that causes birth defects, it is a loss of a relatively small percentage of performance in a video game.
 

kevinsbane

Senior member
Jun 16, 2010
694
0
71
Well, if the whole cost of the card is lost, just send it to me, since it is now worth nothing.

Seriously, you can argue that some performance is lost, and some specs are overstated, but the card is hardly useless, just gives (slightly?) less performance under some circumstances, but even that is very hard to quantify. The most that a buyer can say is "lost" is the difference in performance between the 970 and what a competing card of the same price would have given if you had not bought the card (the 970).
That's assuming I would have bought a card in the first place. The fact of the matter - I was debating whether or not to buy the card - not that I needed it, but I thought it was a good enough GTX 980 alternative for pretty cheap. My alternative? Just use my old card, and not buy a new card. Those were my two courses of action: do nothing, or buy a GTX 970. I didn't need a GTX 970. However, the 4GB/256bit/224GB/s spec, same as the GTX 980, pushed me over the edge, and I chose to buy - thus, NVIDIA's deceptive and misleading specs resulted in a loss of ~$440 (CAD) for me. Did I gain from it? Sure, I got a nice card I suppose... but it wasn't what I thought I was buying. And therein is the crux of the problem. Because if I had known, I'd've saved the $440 and just lived with a slightly slower video card.

I don't say it's useless, because it isn't. It's a great card for what it is. It just isn't the 4GB, 256bit, 224GB/s card that Nvidia advertised and everyone thought it was.
 
Last edited:

DarkKnightDude

Senior member
Mar 10, 2011
981
44
91
I don't think anyone is saying its a useless card, its the basic fact that they lied about the memory and now the ROPS. I find it mindboggling anyone can defend that.
 

Pneumothorax

Golden Member
Nov 4, 2002
1,182
23
81
That's assuming I would have bought a card in the first place. The fact of the matter - I was debating whether or not to buy the card - not that I needed it, but I thought it was a good enough GTX 980 alternative for pretty cheap. My alternative? Just use my old card, and not buy a new card. Those were my two courses of action: do nothing, or buy a GTX 970. I didn't need a GTX 970. However, the 4GB/256bit/224GB/s spec, same as the GTX 980, pushed me over the edge, and I chose to buy - thus, NVIDIA's deceptive and misleading specs resulted in a loss of ~$440 (CAD) for me. Did I gain from it? Sure, I got a nice card I suppose... but it wasn't what I thought I was buying. And therein is the crux of the problem. Because if I had known, I'd've saved the $440 and just lived with a slightly slower video card.

I don't say it's useless, because it isn't. It's a great card for what it is. It just isn't the 4GB, 256bit, 224GB/s card that Nvidia advertised and everyone thought it was.

I was in the same boat as you and sold a perfectly good 780ti to go 970 SLI for a 1600p setup. If I would've known this issue from the beginning I would've just waited for 390x/GM200...
 

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
Ignoring any ethical issues, I also find it odd people are complaining about lost performance. There is no lost performance, as all the benchmarks you've seen in every review, has been done on the same 970's you bought. There may be some problems with the spec's you thought you were getting, but the performance is still the same as what you were promised by all the reviews done.

This isn't an argument about the legality, ethics or anything related to the lawsuit, just there is no lost performance. It is a loss of spec's.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
Interesting that all reviews have virtually stopped on the 970. I say it just shows that the tech press really isn't concerned with giving us the answers. Just selling product for the industry and collecting their paychecks.

He said he would not have bought it. Why ask him again? This is a "read my lips" moment.

very well said. :thumbsup: The tech press in the US is completely not interested in investigating this issue as it would hurt their kickbacks from Nvidia. Few like pcper did frametime testing in single and multi GPU with GTX 970. hardware canucks did avg fps testing on few games. But the others are just not bothered. If any tech site has guts they should do an extensive article with frametime testing on R9 290X and GTX 970 in the latest games which hit more than 3.5 GB VRAM usage at 1440p and 4k. Games like Middle Earth, AC Unity, Lords of the Fallen, COD AW, Evolve are all good candidates.

Even more there is not a single tech site which is condemning the utter lack of CF support in the most recent Gameworks titles like AC Unity, Farcry 4, Dying Light, Evolve. hardocp was vociferous in condemning Ubisoft for FC4 and AC Unity.

http://www.hardocp.com/article/2014/11/21/far_cry_4_video_card_performance_iq_preview/7#.VOq8nOEwDcw

"SLI and CrossFire Support

We would rate Far Cry 4's SLI and CF support below average for Far Cry 4's launch. SLI is supported, but there have been reported some issues.While NVIDIA SLI works somewhat, AMD CrossFire simply does not work at all yet at launch. From AMD's own driver page:
"The AMD CrossFire profile for Far Cry 4 is currently disabled in this driver while AMD works with Ubisoft to investigate an issue where AMD CrossFire configurations are not performing as intended. An update is expected on this issue in the near future through an updated game patch or an AMD driver posting."
With no CrossFire support at launch AMD users with CrossFire setups are left in the dark with but single-GPU performance. This means the money they spent on two or more video cards is wasted right now. This is unacceptable for a high profile game such as this."

http://www.hardocp.com/article/2014..._performance_video_card_review/7#.VOq8xeEwDcw

"SLI and CrossFire are features modern game's should support well at launch. SLI worked, but there were a lot of bugs associated with it at game launch. Many of those bugs have been worked out, and performance is quite good finally after four patches. CrossFire however is broken, and that isn't acceptable."

But even other Gameworks title like Dying Light and Evolve which are developed by other developers have no CF support. On the contrary every Gaming Evolved title like Dragon Age Inquisition, Alien Isolation, Civilization Beyond Earth works perfectly on SLI with good scaling. This is something that needs highlighting and the press should play a role in strongly criticizing Nvidia's Gameworks program and the developer for their slackness in supporting AMD CF.

The GPU market is headed in an unhealthy direction towards a Nvidia monopoly and the biggest losers are going to be the consumers. We can attribute a major part of the blame to AMD(with regards to loss of market share due to inferior products) but Nvidia's tactics are also condemnable. Anyway don't be surprised if Nvidia starts charging USD 2000 for their flagships in 2016 or 2017. D:
 
Last edited:

ITMax1981

Banned
Feb 23, 2015
1
0
0
It's not false advertising. The card has the 4GB on there. It performs great. If nobody had told people this, wouldn't everything have been fine? Who would have been hurt?
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
Ignoring any ethical issues, I also find it odd people are complaining about lost performance. There is no lost performance, as all the benchmarks you've seen in every review, has been done on the same 970's you bought. There may be some problems with the spec's you thought you were getting, but the performance is still the same as what you were promised by all the reviews done.

This isn't an argument about the legality, ethics or anything related to the lawsuit, just there is no lost performance. It is a loss of spec's.

The ethics is the core issue at stake here. Performance is a problem too in cases where VRAM usage hits > 3.5 GB VRAM especially with uneven frametimes even in single GPU at 1440p and 4k but that was actually not highlighted by the tech press as they were busy testing old games and not new ones like Middle Earth, COD AW, Lords of the Fallen, Evolve. As more recent games like Evolve are showing this problem is affecting even single GTX 970 when VRAM usage hits upto 4GB

http://www.techspot.com/review/962-evolve-benchmarks/page4.html

"Disappointingly, the GTX 970 averaged just 44fps, being only marginally faster than the old HD 7970 GHz at 41fps. We believe this massive reduction in performance is due to the GTX 970's partitioned memory configuration."

Anyway in the reviewers favour we can say that they were not aware of the spec misrepresentation and thus did not test to actively evaluate the problems associated when VRAM > 3.5 GB is used in the latest games. But the utter lack of professional testing and evaluation of this problem after it was revealed is disconcerting. pcper was the only US site to do frametime testing. hardwarecanucks did avg fps testing and they saw the GTX 970 taking huge perf hits once VRAM > 3.5 GB was used ing ames like Shadow of Mordor

http://www.pcper.com/reviews/Graphi...ory-Performance/COD-Advanced-Warfare-and-Clos

http://www.pcper.com/reviews/Graphi...ed-Tested-SLI/COD-Advanced-Warfare-and-Closin

http://www.hardwarecanucks.com/foru...68595-gtx-970s-memory-explained-tested-2.html

I expected more from sites like techreport, anandtech, tomshardware on this issue but unfortunately nothing yet. Its been a month from the time the GTX 970 memory partition was revealed. Its very disappointing to see the silence by the tech press on this problem and how it affects consumers especially those that paid USD 700 - USD 800 to buy GTX 970 SLI for 1440p and 4k gaming.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |